An epidemiologic critique of current microbial risk assessment practices: the importance of prevalence and test accuracy data

Data bias in microbial risk assessment, specifically in prevalence estimates, attributable to the use of imperfect tests is discussed. Also, the techniques available to risk analysts for the adjustment of test-based estimates to true prevalence estimates are described. Methods for the evaluation of test accuracy in the absence of a gold standard reference test that provide useful alternative approach to traditional methods are also described. Discussions on the utility of Bayesian methods and Markov chain Monte Carlo simulation for prevalence inferences for rare events and clustered data obtained from single and multiple populations, and for estimation of the accuracy of correlated diagnostic tests are also presented.